Jonathan Parion

Perovskite solar cells with a wide bandgap (WBG) perovskite absorber of 1.68 eV are fabricated and their performance evolution under accelerated stressing conditions are compared with 1.61 eV reference devices. The cells are processed entirely with scalable deposition methods, to guarantee their relevance for industrial application. Different stress tests, following the International Summit on Organic Photovoltaic Stability (ISOS) protocols, are performed, namely prolonged exposure to light (ISOS-L1), heat (ISOS-D2) and a combination of these (ISOS-L2). First, the ISOS-L1 test highlights the excellent stability of the chosen WBG composition, with minimal degradation after 60 h. Secondly, the ISOS-D2 test led to a more significant degradation of the WBG cells, with only 80% efficiency retained after 95 h. The main cause of degradation was found to be interface-related, specifically the formation of a charge transport barrier at the perovskite/electron transport layer interface, while the perovskite absorption properties remained unaffected by the stress test. Finally, the ISOS-L2 test led to an even faster degradation, with only 80% efficiency retained after 35 h. There, the perovskite absorber itself was found to be significantly degraded due to the combined action of light and heat. Altogether, this study highlights the main degradation pathways in WBG perovskite cells while showing the importance of diversified and combined stresses in evaluating their stability.

Perovskite solar cells are one of the most promising photovoltaic technologies in the last decades. Inverted (p–i–n) cells using NiO_X as hole‐transport layer (HTL) have gained attention due to their easy fabrication methods and high stability, although they often exhibit reduced efficiencies due to non onlyoptimized energy‐level alignment. To address this issue, different approaches have been developed, such as the use of self‐assembled monolayers (SAMs) on top of the HTL. Herein, a comparative study between regular p–i–n cells and cells using Me‐PACz as an SAM on top of NiO_X is we presented. Devices with SAM exhibit enhanced open‐circuit voltage and efficiency. Temperature DC and AC characterization reveals that the incorporation of SAM reduces recombination at the interface, as seen from the comparison of the perovskite bandgap (1.6 eV) and carrier activation energy ≈1.1 and ≈1.59 eV for reference and SAM, respectively. Finally, an outdoor degradation experiment with minimodules has been conducted. The experiment spanned for more than 500 days, and results show that minimodules with SAM were less stable than those based on the reference layer structure. This is due to a severe decrease in the short‐circuit current, which could be attributed to a deterioration of the SAM spacer.

Understanding the behavior of mobile ions in perovskite‐based solar cells (PSCs) is crucial for improving their performance and stability, which belong to the key hurdles in advancing this technology toward commercialization. This study explores the role of mobile ions in PSCs using the comprehensive technology computer‐aided design model which is extended to simulate the frequency‐dependent capacitance (C–f) of PSCs. It is compared with equivalent circuit approaches showcasing the validity and advantages of full device modeling. By combining the simulation of full measurement procedures with C–f and J–V measurements on experimental test structures, the observed C–f characteristics can be quantitatively related to performance losses in scan‐time‐dependent J–V curves, both originating from ion diffusion. With this combined analysis, insights can be provided on the physical origin and interpretation of the different C–f plateaus caused by the displacement of ions. Finally, the C–f characteristics of PSCs under illumination and the impact of band alignment and recombination at the perovskite interface are investigated. Experimental PSCs with and without electron‐transport layer passivation are fabricated, showing a good agreement between the simulated and measured C–f and pointing toward a lower surface recombination for the passivated PSC. This study shows how drift‐diffusion modeling helps to characterize and interpret capacitance‐based data.

A multifaceted characterization approach is proposed, aiming to establish a link between nanoscale electrical properties and macroscale device characteristics. Current–voltage (I–V) measurements are combined with admittance spectroscopy (AS) and deep‐level transient spectroscopy (DLTS) for the analysis of charge‐related performance losses with time‐of‐flight secondary‐ion mass spectrometry to complete the understanding of ionic motion in the device. This is applied to the study of surface treatment in perovskite solar cells, which implements several strategies to improve band alignment, perovskite grain growth, and chemical passivation. An increase of both open‐circuit voltage (V_oc) and fill factor of respectively 90 mV and 11% is shown after surface treatment, with an absolute efficiency increase of 4%. AS measurements, coupled with a lumped elements model, rule out the impact of transport layers as the origin of the performance improvement, rather pointing toward a reduction in ionic resistance in the perovskite bulk. Analysis of the DLTS response yields an activation energy of 0.41 eV, which is likely related to the same ionic mechanism discovered with AS. Finally, both of these techniques enable to show that the surface treatment main contribution is to reduce ion‐related recombination of charge carriers.

Lithium fluoride (LiF) is currently a very popular dielectric material used as a passivation or transport layer in a variety of applications, especially in high-efficiency solar cells. Despite this, its conduction properties and interface behavior with silicon remain largely unexplored. In this work, a LiF metal–insulator–semiconductor (MIS) structure is fabricated and characterized, and its properties are compared to the well-understood aluminum oxide (Al2O3) MIS structure. First, a higher current density in LiF compared to Al2O3 is highlighted, as well as its PN junction-like behavior with n-type silicon (n-Si), being rather unconventional for a dielectric layer. C–V measurements showcase the likely presence of an interface defect, causing an increase in the apparent doping and a shift in the flatband voltage VFB by +70 meV. This defect is found to be of the acceptor type, which renders the interface fixed charge more negative and improves the field-effect passivation in the case of a negative Qf. Finally, a density of interface states Dit≈2×1011 cm−2 eV−1 was found for LiF/n-Si, which is a low value showing appropriate chemical passivation at the interface. Overall, this work enables us to shed more light on the interface properties of LiF on n-Si, which is an essential step toward its wider use in state-of-the-art solar cells and other silicon-based devices.